This film optical recording layers using chalcogenide thin-films and amorphous to crystalline phase transitions have been the subject of many investigations since the early 1970's. The initial interests were focused on "erasable", and therefore reusable, optical recording layers since the amorphous to crystalline transition is, in principle, a reversible process. Such layers are generally prepared by a vacuum process. The layer is amorphous when so prepared. A low power, relatively long duration laser pulse is used to heat a local spot on the layer to below the melting point for a sufficient length of time to cause the spots to crystallize. These crystalline spots can in turn be heated, by a higher power, shorter duration laser, above the melting point of the crystallized spots to randomize the structure of the spots. The layer is designed such that upon the termination of the laser pulse the cooling rate of the heated spot is high enough that the randomized structure is frozen to achieve an amorphous state.
Thus by adjusting the laser power and duration, the state of a selected area on the layer can be switched between the amorphous state and the crystalline state to create a pattern of amorphous and crystalline spots which can be used for information storage. Since the phase transition is reversible, the pattern can be erased and replaced with a different recorded pattern. Theoretically, this erase-write cycle can be carried out any number of times.
A principal difficulty is that the rate of crystallization of most layers studied is usually too low. For practical applications, it is desirable to have layers which can be crystallized by laser pulses shorter than a microsecond (.mu.s). Presently, few materials have demonstrated such capabilities. For some materials with high crystallization rates (e.g. Te-Sn alloy), the data retention times are often not adequate because of the instability of the amorphous state.
Because of the slow crystallization of most materials, the crystallization step is generally used as the erasure step in erasable optical recording layers. A laser spot elongated in the direction of the laser movement is used to give an effectively long duration laser exposure. Such long laser spots cannot be used for high density recordings. The amorphizing step, on the other hand, is used as the recording step since this can be achieved with short laser pulse, and hence can be done at high speed.
Very few materials are known for optical recording layers in which the above described write-erase-write cycle is of practical use. No erasable phase-change type optical recording layers have been commercialized.
A good deal of attention has also focused on so-called "write-once" thin film optical recording layers. Write-once simply means that the layers can be recorded upon only once. Such layers cannot be erased and reused for a subsequent recording.
Since thin film optical recording layers are generally amorphous when prepared, it is desirable to use the crystallization step as the recording step in write-once layers. However, the problem of slow crystallization prevents the achievement of high data rates. High data rates are critical for write-once layers designed for use with computers.
European Patent Application No. 0184452 broadly discloses erasable optical recording layers of antimony-indium and antimony-indium-tin alloys. Information recording and erasure are said to be achieved by switching the layers between two different crystalline states. The layers are generally prepared in the amorphous states which have to be first converted into one of the two crystalline states before information can be recorded. The crystallization step, achieved by either a bulk heat-treatment or a prolonged laser exposure, is said to have a lower reflectance than the amorphous state. The examples indicate that the materials disclosed therein have a very slow rate of crystallization. This application further teaches that the optical recording layers disclosed therein are unsuitable for use in the amorphous-to-crystalline transition mechanism because of the instability of the amorphous state in general.
Experimental evidence has shown that the crystalline-crystalline recordings and the fast amorphous-to-crystalline recordings are mutually exclusive. Compositions which demonstrate properties suitable for one mode of recording are not suitable for the other mode of recording. Moreover, many compositions within the broad disclosure of European Patent No. 0184452 are incapable of forming write-once recording layers; require commercially unacceptable amounts of laser power; have poor dynamic carrier to noise ratios (CNR); have poor dynamic sensitivity; or the films are crystalline when first formed; or one of the crystalline phases formed is not stable.
Another problem is that many of the chalcogen containing materials which undergo the amorphous-to-crystalline transition mechanism are usually corrosion prone.
The problem is that the prior art has not provided write-once optical recording layers which possess the combination of (a) a crystallization rate less than 1.0 .mu.s, (b) good corrosion resistance, (c) a stable amorphous state and (d) a capability of high rate, high density recordings.